Please use this identifier to cite or link to this item:
Title: Fully PolSAR image classification using machine learning techniques and reaction-diffusion systems
Authors: Gomez, Luis 
Alvarez, Luis 
Mazorra Aguiar, Luis
Frery, Alejandro C. 
UNESCO Clasification: 220990 Tratamiento digital. Imágenes
120601 Construcción de algoritmos
120326 Simulación
120602 Ecuaciones diferenciales
Keywords: Image processing
Image analysis
SAR polarimetry
Issue Date: 2017
Journal: Neurocomputing 
Abstract: In this paper, we study the problem of supervised Fully PolSAR (polarimetric synthetic aperture radar) image classification. We estimate a complex Wishart model distribution for each class using training data, and we use such models to design a new classification procedure based on a diffusion-reaction equation. The method relies on simultaneously filtering and classifying pixels within the image. The diffusion term smooths the patches within the image, and the reaction term tends to move the pixel values towards the closest (in the sense of stochastic distances) representative class. We present a detailed study of the method accuracy using both simulated and true data, and we provide optimum parameters for its use. We show that the proposed method outperforms the results obtained using maximum likelihood and usual stochastic distance classification methods.
ISSN: 0925-2312
DOI: 10.1016/j.neucom.2016.08.140
Source: Neurocomputing[ISSN 0925-2312],v. 255, p. 52-60
Appears in Collections:Artículos
Show full item record

Google ScholarTM




Export metadata

Items in accedaCRIS are protected by copyright, with all rights reserved, unless otherwise indicated.